-We as Instrumentalists used CNN model to classify sound files on Urban Sound Dataset. There will be two notebooks in this repo in which one contains creating spectrograms and preprocessing them; since it took time we used pre-processed item set that you provided with us. It's there so that you can see we are able to do it. The second notebook actually directly uses the spectrograms and create the sets to train our model.
- Urban Sound dataset contains 8732 labeled of urban sounds from 10 classes: air_conditioner, car_horn, children_playing, dog_bark, drilling, engine_idling, gun_shot, jackhammer, siren, and street_music. All output of our model can be found in the model notebook.
-For more info about the dataset checkout link below;